Optimising the definition of the target volume for lung cancer radiotherapy

Radiotherapy, with or without chemotherapy, has an important role in the management of lung cancer. Radiotherapy involves the precise delivery of ionising radiation to the tumour. In order to deliver the treatment accurately, the clinician must precisely outline the gross tumour volume (GTV), mostly on computed tomography (CT) images. However, due to poor soft-tissue contrast, it can be difficult to distinguish the tumour from adjacent tissues leading to large interobserver variation and differences in interpretation. GTV definition has often been described as the weakest link in radiotherapy with its inaccuracy potentially leading to missing the tumour or unnecessarily irradiating normal tissue. Therefore, the general aim of this thesis was to optimise the definition of the GTV by improving simple autosegmentation tools, improving image quality, and developing new training and quality assurance (QA) methods.
The findings of the thesis indicate that to date, it is still not possible to eliminate interobserver variation in the definition of the GTV via a single intervention. Improvements in image quality can reduce interobserver variations in some cases. Simple adaptive threshold segmentation based on tumour size provided segmentation in good agreement with pathology data. However, these interventions may not necessarily reduce image interpretation differences that ultimately cause the largest interobserver variation. Training, interprofessional collaboration, and QA checks are essential to address this challenge. Clinical workflows need to be further developed to facilitate the introduction of interprofessional QA checks in clinical practice and to better integrate training on target volume definition within the QA programme.